[thirdparty/git.git] / git-svnimport.perl

#!/usr/bin/perl -w

# This tool is copyright (c) 2005, Matthias Urlichs.
# It is released under the Gnu Public License, version 2.
#
# The basic idea is to pull and analyze SVN changes.
#
# Checking out the files is done by a single long-running CVS connection
# / server process.
#
# The head revision is on branch "origin" by default.
# You can change that with the '-o' option.

require v5.8.0; # for shell-safe open("-|",LIST)
use strict;
use warnings;
use Getopt::Std;
use File::Spec;
use File::Temp qw(tempfile);
use File::Path qw(mkpath);
use File::Basename qw(basename dirname);
use Time::Local;
use IO::Pipe;
use POSIX qw(strftime dup2);
use IPC::Open2;
use SVN::Core;
use SVN::Ra;

die "Need CVN:COre 1.2.1 or better" if $SVN::Core::VERSION lt "1.2.1";

$SIG{'PIPE'}="IGNORE";
$ENV{'TZ'}="UTC";

our($opt_h,$opt_o,$opt_v,$opt_u,$opt_C,$opt_i,$opt_m,$opt_M,$opt_t,$opt_T,$opt_b,$opt_s,$opt_l);

sub usage() {
	print STDERR <<END;
Usage: ${\basename $0}     # fetch/update GIT from CVS
       [-o branch-for-HEAD] [-h] [-v] [-l max_num_changes]
       [-C GIT_repository] [-t tagname] [-T trunkname] [-b branchname]
       [-i] [-u] [-s start_chg] [-m] [-M regex] [SVN_URL]
END
	exit(1);
}

getopts("b:C:hil:mM:o:s:t:T:uv") or usage();
usage if $opt_h;

my $tag_name = $opt_t || "tags";
my $trunk_name = $opt_T || "trunk";
my $branch_name = $opt_b || "branches";

@ARGV <= 1 or usage();

$opt_o ||= "origin";
$opt_l = 100 unless defined $opt_l;
my $git_tree = $opt_C;
$git_tree ||= ".";

my $cvs_tree;
if ($#ARGV == 0) {
	$cvs_tree = $ARGV[0];
} elsif (-f 'CVS/Repository') {
	open my $f, '<', 'CVS/Repository' or 
	    die 'Failed to open CVS/Repository';
	$cvs_tree = <$f>;
	chomp $cvs_tree;
	close $f;
} else {
	usage();
}

our @mergerx = ();
if ($opt_m) {
	@mergerx = ( qr/\W(?:from|of|merge|merging|merged) (\w+)/i );
}
if ($opt_M) {
	push (@mergerx, qr/$opt_M/);
}

select(STDERR); $|=1; select(STDOUT);


package SVNconn;
# Basic SVN connection.
# We're only interested in connecting and downloading, so ...

use File::Spec;
use File::Temp qw(tempfile);
use POSIX qw(strftime dup2);

sub new {
	my($what,$repo) = @_;
	$what=ref($what) if ref($what);

	my $self = {};
	$self->{'buffer'} = "";
	bless($self,$what);

	$repo =~ s#/+$##;
	$self->{'fullrep'} = $repo;
	$self->conn();

	return $self;
}

sub conn {
	my $self = shift;
	my $repo = $self->{'fullrep'};
	my $s = SVN::Ra->new($repo);

	die "SVN connection to $repo: $!\n" unless defined $s;
	$self->{'svn'} = $s;
	$self->{'repo'} = $repo;
	$self->{'maxrev'} = $s->get_latest_revnum();
}

sub file {
	my($self,$path,$rev) = @_;
	my $res;

	my ($fh, $name) = tempfile('gitsvn.XXXXXX', 
		    DIR => File::Spec->tmpdir(), UNLINK => 1);

	print "... $rev $path ...\n" if $opt_v;
	eval { $self->{'svn'}->get_file($path,$rev,$fh); };
	if ($@ and $@ !~ /Attempted to get checksum/) {
	    # retry
	    $self->conn();
		eval { $self->{'svn'}->get_file($path,$rev,$fh); };
	};
	return () if $@ and $@ !~ /Attempted to get checksum/;
	die $@ if $@;
	close ($fh);

	return ($name, $res);
}


package main;

my $svn = SVNconn->new($cvs_tree);


sub pdate($) {
	my($d) = @_;
	$d =~ m#(\d\d\d\d)-(\d\d)-(\d\d)T(\d\d):(\d\d):(\d\d)#
		or die "Unparseable date: $d\n";
	my $y=$1; $y-=1900 if $y>1900;
	return timegm($6||0,$5,$4,$3,$2-1,$y);
}

sub getwd() {
	my $pwd = `pwd`;
	chomp $pwd;
	return $pwd;
}


sub get_headref($$) {
    my $name    = shift;
    my $git_dir = shift; 
    my $sha;
    
    if (open(C,"$git_dir/refs/heads/$name")) {
	chomp($sha = <C>);
	close(C);
	length($sha) == 40
	    or die "Cannot get head id for $name ($sha): $!\n";
    }
    return $sha;
}


-d $git_tree
	or mkdir($git_tree,0777)
	or die "Could not create $git_tree: $!";
chdir($git_tree);

my $orig_branch = "";
my $forward_master = 0;
my %branches;

my $git_dir = $ENV{"GIT_DIR"} || ".git";
$git_dir = getwd()."/".$git_dir unless $git_dir =~ m#^/#;
$ENV{"GIT_DIR"} = $git_dir;
my $orig_git_index;
$orig_git_index = $ENV{GIT_INDEX_FILE} if exists $ENV{GIT_INDEX_FILE};
my ($git_ih, $git_index) = tempfile('gitXXXXXX', SUFFIX => '.idx',
				    DIR => File::Spec->tmpdir());
close ($git_ih);
$ENV{GIT_INDEX_FILE} = $git_index;
my $maxnum = 0;
my $last_rev = "";
my $last_branch;
my $current_rev = $opt_s ? ($opt_s-1) : 0;
unless(-d $git_dir) {
	system("git-init-db");
	die "Cannot init the GIT db at $git_tree: $?\n" if $?;
	system("git-read-tree");
	die "Cannot init an empty tree: $?\n" if $?;

	$last_branch = $opt_o;
	$orig_branch = "";
} else {
	-f "$git_dir/refs/heads/$opt_o"
		or die "Branch '$opt_o' does not exist.\n".
		       "Either use the correct '-o branch' option,\n".
		       "or import to a new repository.\n";

	-f "$git_dir/svn2git"
		or die "'$git_dir/svn2git' does not exist.\n".
		       "You need that file for incremental imports.\n";
	$last_branch = basename(readlink("$git_dir/HEAD"));
	unless($last_branch) {
		warn "Cannot read the last branch name: $! -- assuming 'master'\n";
		$last_branch = "master";
	}
	$orig_branch = $last_branch;
	$last_rev = get_headref($orig_branch, $git_dir);
	if (-f "$git_dir/SVN2GIT_HEAD") {
		die <<EOM;
SVN2GIT_HEAD exists.
Make sure your working directory corresponds to HEAD and remove SVN2GIT_HEAD.
You may need to run

    git-read-tree -m -u SVN2GIT_HEAD HEAD
EOM
	}
	system('cp', "$git_dir/HEAD", "$git_dir/SVN2GIT_HEAD");

	$forward_master =
	    $opt_o ne 'master' && -f "$git_dir/refs/heads/master" &&
	    system('cmp', '-s', "$git_dir/refs/heads/master", 
				"$git_dir/refs/heads/$opt_o") == 0;

	# populate index
	system('git-read-tree', $last_rev);
	die "read-tree failed: $?\n" if $?;

	# Get the last import timestamps
	open my $B,"<", "$git_dir/svn2git";
	while(<$B>) {
		chomp;
		my($num,$branch,$ref) = split;
		$branches{$branch}{$num} = $ref;
		$branches{$branch}{"LAST"} = $ref;
		$current_rev = $num if $current_rev < $num;
	}
	close($B);
}
-d $git_dir
	or die "Could not create git subdir ($git_dir).\n";

open BRANCHES,">>", "$git_dir/svn2git";


## cvsps output:
#---------------------
#PatchSet 314
#Date: 1999/09/18 13:03:59
#Author: wkoch
#Branch: STABLE-BRANCH-1-0
#Ancestor branch: HEAD
#Tag: (none)
#Log:
#    See ChangeLog: Sat Sep 18 13:03:28 CEST 1999  Werner Koch
#Members:
#	README:1.57->1.57.2.1
#	VERSION:1.96->1.96.2.1
#
#---------------------

my $state = 0;

sub get_file($$$) {
	my($rev,$branch,$path) = @_;

	# revert split_path(), below
	my $svnpath;
	$path = "" if $path eq "/"; # this should not happen, but ...
	if($branch eq "/") {
		$svnpath = "/$trunk_name/$path";
	} elsif($branch =~ m#^/#) {
		$svnpath = "/$tag_name$branch/$path";
	} else {
		$svnpath = "/$branch_name/$branch/$path";
	}

	# now get it
	my ($name, $res) = eval { $svn->file($svnpath,$rev); };
	return () unless defined $name;

	open my $F, '-|', "git-hash-object", "-w", $name
		or die "Cannot create object: $!\n";
	my $sha = <$F>;
	chomp $sha;
	close $F;
	my $mode = "0644"; # SV does not seem to store any file modes
	return [$mode, $sha, $path];
}

sub split_path($$) {
	my($rev,$path) = @_;
	my $branch;

	if($path =~ s#^/\Q$tag_name\E/([^/]+)/?##) {
		$branch = "/$1";
	} elsif($path =~ s#^/\Q$trunk_name\E/?##) {
		$branch = "/";
	} elsif($path =~ s#^/\Q$branch_name\E/([^/]+)/?##) {
		$branch = $1;
	} else {
		print STDERR "$rev: Unrecognized path: $path\n";
		return ()
	}
	$path = "/" if $path eq "";
	return ($branch,$path);
}

sub commit {
	my($branch, $changed_paths, $revision, $author, $date, $message) = @_;
	my($author_name,$author_email,$dest);
	my(@old,@new);

	if (not defined $author) {
		$author_name = $author_email = "unknown";
	} elsif ($author =~ /^(.*?)\s+<(.*)>$/) {
		($author_name, $author_email) = ($1, $2);
	} else {
		$author =~ s/^<(.*)>$/$1/;
		$author_name = $author_email = $author;
	}
	$date = pdate($date);

	my $tag;
	my $parent;
	if($branch eq "/") { # trunk
		$parent = $opt_o;
	} elsif($branch =~ m#^/(.+)#) { # tag
		$tag = 1;
		$parent = $1;
	} else { # "normal" branch
		# nothing to do
		$parent = $branch;
	}
	$dest = $parent;

	my $prev = $changed_paths->{"/"};
	if($prev and $prev->[0] eq "A") {
		delete $changed_paths->{"/"};
		my $oldpath = $prev->[1];
		my $rev;
		if(defined $oldpath) {
			my $p;
			($parent,$p) = split_path($revision,$oldpath);
			if($parent eq "/") {
				$parent = $opt_o;
			} else {
				$parent =~ s#^/##; # if it's a tag
			}
		} else {
			$parent = undef;
		}
	}

	my $rev;
	if($revision > $opt_s and defined $parent) {
		open(H,"git-rev-parse --verify $parent |");
		$rev = <H>;
		close(H) or do {
			print STDERR "$revision: cannot find commit '$parent'!\n";
			return;
		};
		chop $rev;
		if(length($rev) != 40) {
			print STDERR "$revision: cannot find commit '$parent'!\n";
			return;
		}
		$rev = $branches{($parent eq $opt_o) ? "/" : $parent}{"LAST"};
		if($revision != $opt_s and not $rev) {
			print STDERR "$revision: do not know ancestor for '$parent'!\n";
			return;
		}
	} else {
		$rev = undef;
	}

#	if($prev and $prev->[0] eq "A") {
#		if(not $tag) {
#			unless(open(H,"> $git_dir/refs/heads/$branch")) {
#				print STDERR "$revision: Could not create branch $branch: $!\n";
#				$state=11;
#				next;
#			}
#			print H "$rev\n"
#				or die "Could not write branch $branch: $!";
#			close(H)
#				or die "Could not write branch $branch: $!";
#		}
#	}
	if(not defined $rev) {
		unlink($git_index);
	} elsif ($rev ne $last_rev) {
		print "Switching from $last_rev to $rev ($branch)\n" if $opt_v;
		system("git-read-tree", $rev);
		die "read-tree failed for $rev: $?\n" if $?;
		$last_rev = $rev;
	}

	my $cid;
	if($tag and not %$changed_paths) {
		$cid = $rev;
	} else {
		while(my($path,$action) = each %$changed_paths) {
			if ($action->[0] eq "A") {
				my $f = get_file($revision,$branch,$path);
				push(@new,$f) if $f;
			} elsif ($action->[0] eq "D") {
				push(@old,$path);
			} elsif ($action->[0] eq "M") {
				my $f = get_file($revision,$branch,$path);
				push(@new,$f) if $f;
			} elsif ($action->[0] eq "R") {
				# refer to a file/tree in an earlier commit
				push(@old,$path); # remove any old stuff

				# ... and add any new stuff
				my($b,$p) = split_path($revision,$action->[1]);
				open my $F,"-|","git-ls-tree","-r","-z", $branches{$b}{$action->[2]}, $p;
				local $/ = '\0';
				while(<$F>) {
					chomp;
					my($m,$p) = split(/\t/,$_,2);
					my($mode,$type,$sha1) = split(/ /,$m);
					next if $type ne "blob";
					push(@new,[$mode,$sha1,$p]);
				}
			} else {
				die "$revision: unknown action '".$action->[0]."' for $path\n";
			}
		}

		if(@old) {
			open my $F, "-|", "git-ls-files", "-z", @old or die $!;
			@old = ();
			local $/ = '\0';
			while(<$F>) {
				chomp;
				push(@old,$_);
			}
			close($F);

			while(@old) {
				my @o2;
				if(@old > 55) {
					@o2 = splice(@old,0,50);
				} else {
					@o2 = @old;
					@old = ();
				}
				system("git-update-index","--force-remove","--",@o2);
				die "Cannot remove files: $?\n" if $?;
			}
		}
		while(@new) {
			my @n2;
			if(@new > 12) {
				@n2 = splice(@new,0,10);
			} else {
				@n2 = @new;
				@new = ();
			}
			system("git-update-index","--add",
				(map { ('--cacheinfo', @$_) } @n2));
			die "Cannot add files: $?\n" if $?;
		}

		my $pid = open(C,"-|");
		die "Cannot fork: $!" unless defined $pid;
		unless($pid) {
			exec("git-write-tree");
			die "Cannot exec git-write-tree: $!\n";
		}
		chomp(my $tree = <C>);
		length($tree) == 40
			or die "Cannot get tree id ($tree): $!\n";
		close(C)
			or die "Error running git-write-tree: $?\n";
		print "Tree ID $tree\n" if $opt_v;

		my $pr = IO::Pipe->new() or die "Cannot open pipe: $!\n";
		my $pw = IO::Pipe->new() or die "Cannot open pipe: $!\n";
		$pid = fork();
		die "Fork: $!\n" unless defined $pid;
		unless($pid) {
			$pr->writer();
			$pw->reader();
			open(OUT,">&STDOUT");
			dup2($pw->fileno(),0);
			dup2($pr->fileno(),1);
			$pr->close();
			$pw->close();

			my @par = ();
			@par = ("-p",$rev) if defined $rev;

			# loose detection of merges
			# based on the commit msg
			foreach my $rx (@mergerx) {
				if ($message =~ $rx) {
					my $mparent = $1;
					if ($mparent eq 'HEAD') { $mparent = $opt_o };
					if ( -e "$git_dir/refs/heads/$mparent") {
						$mparent = get_headref($mparent, $git_dir);
						push @par, '-p', $mparent;
						print OUT "Merge parent branch: $mparent\n" if $opt_v;
					}
				} 
			}

			exec("env",
				"GIT_AUTHOR_NAME=$author_name",
				"GIT_AUTHOR_EMAIL=$author_email",
				"GIT_AUTHOR_DATE=".strftime("+0000 %Y-%m-%d %H:%M:%S",gmtime($date)),
				"GIT_COMMITTER_NAME=$author_name",
				"GIT_COMMITTER_EMAIL=$author_email",
				"GIT_COMMITTER_DATE=".strftime("+0000 %Y-%m-%d %H:%M:%S",gmtime($date)),
				"git-commit-tree", $tree,@par);
			die "Cannot exec git-commit-tree: $!\n";
		}
		$pw->writer();
		$pr->reader();

		$message =~ s/[\s\n]+\z//;

		print $pw "$message\n"
			or die "Error writing to git-commit-tree: $!\n";
		$pw->close();

		print "Committed change $revision:$branch ".strftime("%Y-%m-%d %H:%M:%S",gmtime($date)).")\n" if $opt_v;
		chomp($cid = <$pr>);
		length($cid) == 40
			or die "Cannot get commit id ($cid): $!\n";
		print "Commit ID $cid\n" if $opt_v;
		$pr->close();

		waitpid($pid,0);
		die "Error running git-commit-tree: $?\n" if $?;
	}

	if(not defined $dest) {
		print "... no known parent\n" if $opt_v;
	} elsif(not $tag) {
		print "Writing to refs/heads/$dest\n" if $opt_v;
		open(C,">$git_dir/refs/heads/$dest") and 
		print C ("$cid\n") and
		close(C)
			or die "Cannot write branch $dest for update: $!\n";
	}

	if($tag) {
		my($in, $out) = ('','');
		$last_rev = "-" if %$changed_paths;
		# the tag was 'complex', i.e. did not refer to a "real" revision
		
		$dest =~ tr/_/\./ if $opt_u;

		my $pid = open2($in, $out, 'git-mktag');
		print $out ("object $cid\n".
		    "type commit\n".
		    "tag $dest\n".
		    "tagger $author_name <$author_email>\n") and
		close($out)
		    or die "Cannot create tag object $dest: $!\n";

		my $tagobj = <$in>;
		chomp $tagobj;

		if ( !close($in) or waitpid($pid, 0) != $pid or
				$? != 0 or $tagobj !~ /^[0123456789abcdef]{40}$/ ) {
			die "Cannot create tag object $dest: $!\n";
		}

		open(C,">$git_dir/refs/tags/$dest") and
		print C ("$tagobj\n") and
		close(C)
			or die "Cannot create tag $branch: $!\n";

		print "Created tag '$dest' on '$branch'\n" if $opt_v;
	}
	$branches{$branch}{"LAST"} = $cid;
	$branches{$branch}{$revision} = $cid;
	$last_rev = $cid;
	print BRANCHES "$revision $branch $cid\n";
	print "DONE: $revision $dest $cid\n" if $opt_v;
}

my ($changed_paths, $revision, $author, $date, $message, $pool) = @_;
sub _commit_all {
	($changed_paths, $revision, $author, $date, $message, $pool) = @_;
	my %p;
	while(my($path,$action) = each %$changed_paths) {
		$p{$path} = [ $action->action,$action->copyfrom_path, $action->copyfrom_rev ];
	}
	$changed_paths = \%p;
}

sub commit_all {
	my %done;
	my @col;
	my $pref;
	my $branch;

	while(my($path,$action) = each %$changed_paths) {
		($branch,$path) = split_path($revision,$path);
		next if not defined $branch;
		$done{$branch}{$path} = $action;
	}
	while(($branch,$changed_paths) = each %done) {
		commit($branch, $changed_paths, $revision, $author, $date, $message);
	}
}

while(++$current_rev < $svn->{'maxrev'}) {
	$svn->{'svn'}->get_log("/",$current_rev,$current_rev,$current_rev,1,1,\&_commit_all,"");
	commit_all();
	if($opt_l and not --$opt_l) {
		print STDERR "Exiting due to a memory leak. Repeat, please.\n";
		last;
	}
}


unlink($git_index);

if (defined $orig_git_index) {
	$ENV{GIT_INDEX_FILE} = $orig_git_index;
} else {
	delete $ENV{GIT_INDEX_FILE};
}

# Now switch back to the branch we were in before all of this happened
if($orig_branch) {
	print "DONE\n" if $opt_v and (not defined $opt_l or $opt_l > 0);
	system("cp","$git_dir/refs/heads/$opt_o","$git_dir/refs/heads/master")
		if $forward_master;
	unless ($opt_i) {
		system('git-read-tree', '-m', '-u', 'SVN2GIT_HEAD', 'HEAD');
		die "read-tree failed: $?\n" if $?;
	}
} else {
	$orig_branch = "master";
	print "DONE; creating $orig_branch branch\n" if $opt_v and (not defined $opt_l or $opt_l > 0);
	system("cp","$git_dir/refs/heads/$opt_o","$git_dir/refs/heads/master")
		unless -f "$git_dir/refs/heads/master";
	unlink("$git_dir/HEAD");
	symlink("refs/heads/$orig_branch","$git_dir/HEAD");
	unless ($opt_i) {
		system('git checkout');
		die "checkout failed: $?\n" if $?;
	}
}
unlink("$git_dir/SVN2GIT_HEAD");
close(BRANCHES);
Commit	Line	Data
eaf718f3 MU	1	#!/usr/bin/perl -w
	2
	3	# This tool is copyright (c) 2005, Matthias Urlichs.
	4	# It is released under the Gnu Public License, version 2.
	5	#
	6	# The basic idea is to pull and analyze SVN changes.
	7	#
	8	# Checking out the files is done by a single long-running CVS connection
	9	# / server process.
	10	#
	11	# The head revision is on branch "origin" by default.
	12	# You can change that with the '-o' option.
	13
	14	require v5.8.0; # for shell-safe open("-\|",LIST)
	15	use strict;
	16	use warnings;
	17	use Getopt::Std;
	18	use File::Spec;
	19	use File::Temp qw(tempfile);
	20	use File::Path qw(mkpath);
	21	use File::Basename qw(basename dirname);
	22	use Time::Local;
	23	use IO::Pipe;
	24	use POSIX qw(strftime dup2);
	25	use IPC::Open2;
	26	use SVN::Core;
	27	use SVN::Ra;
	28
37dcf6de MU	29	die "Need CVN:COre 1.2.1 or better" if $SVN::Core::VERSION lt "1.2.1";
37dcf6de MU	30
eaf718f3 MU	31	$SIG{'PIPE'}="IGNORE";
	32	$ENV{'TZ'}="UTC";
	33
3ef378a6	34	our($opt_h,$opt_o,$opt_v,$opt_u,$opt_C,$opt_i,$opt_m,$opt_M,$opt_t,$opt_T,$opt_b,$opt_s,$opt_l);
eaf718f3 MU	35
	36	sub usage() {
	37	print STDERR <<END;
	38	Usage: ${\basename $0} # fetch/update GIT from CVS
3ef378a6	39	[-o branch-for-HEAD] [-h] [-v] [-l max_num_changes]
eaf718f3	40	[-C GIT_repository] [-t tagname] [-T trunkname] [-b branchname]
7ee74a99	41	[-i] [-u] [-s start_chg] [-m] [-M regex] [SVN_URL]
eaf718f3 MU	42	END
	43	exit(1);
	44	}
	45
3ef378a6	46	getopts("b:C:hil:mM:o:s:t:T:uv") or usage();
eaf718f3 MU	47	usage if $opt_h;
	48
	49	my $tag_name = $opt_t \|\| "tags";
	50	my $trunk_name = $opt_T \|\| "trunk";
	51	my $branch_name = $opt_b \|\| "branches";
	52
	53	@ARGV <= 1 or usage();
	54
	55	$opt_o \|\|= "origin";
3ef378a6	56	$opt_l = 100 unless defined $opt_l;
eaf718f3 MU	57	my $git_tree = $opt_C;
	58	$git_tree \|\|= ".";
	59
	60	my $cvs_tree;
	61	if ($#ARGV == 0) {
	62	$cvs_tree = $ARGV[0];
	63	} elsif (-f 'CVS/Repository') {
	64	open my $f, '<', 'CVS/Repository' or
	65	die 'Failed to open CVS/Repository';
	66	$cvs_tree = <$f>;
	67	chomp $cvs_tree;
	68	close $f;
	69	} else {
	70	usage();
	71	}
	72
	73	our @mergerx = ();
	74	if ($opt_m) {
	75	@mergerx = ( qr/\W(?:from\|of\|merge\|merging\|merged) (\w+)/i );
	76	}
	77	if ($opt_M) {
	78	push (@mergerx, qr/$opt_M/);
	79	}
	80
	81	select(STDERR); $\|=1; select(STDOUT);
	82
	83
	84	package SVNconn;
	85	# Basic SVN connection.
	86	# We're only interested in connecting and downloading, so ...
	87
	88	use File::Spec;
	89	use File::Temp qw(tempfile);
	90	use POSIX qw(strftime dup2);
	91
	92	sub new {
	93	my($what,$repo) = @_;
	94	$what=ref($what) if ref($what);
	95
	96	my $self = {};
	97	$self->{'buffer'} = "";
	98	bless($self,$what);
	99
	100	$repo =~ s#/+$##;
	101	$self->{'fullrep'} = $repo;
	102	$self->conn();
	103
eaf718f3 MU	104	return $self;
	105	}
	106
	107	sub conn {
	108	my $self = shift;
	109	my $repo = $self->{'fullrep'};
	110	my $s = SVN::Ra->new($repo);
	111
	112	die "SVN connection to $repo: $!\n" unless defined $s;
	113	$self->{'svn'} = $s;
	114	$self->{'repo'} = $repo;
	115	$self->{'maxrev'} = $s->get_latest_revnum();
	116	}
	117
	118	sub file {
	119	my($self,$path,$rev) = @_;
	120	my $res;
	121
	122	my ($fh, $name) = tempfile('gitsvn.XXXXXX',
	123	DIR => File::Spec->tmpdir(), UNLINK => 1);
	124
2b5e63d1	125	print "... $rev $path ...\n" if $opt_v;
f02b3eba	126	eval { $self->{'svn'}->get_file($path,$rev,$fh); };
8cd4177d	127	if ($@ and $@ !~ /Attempted to get checksum/) {
eaf718f3 MU	128	# retry
eaf718f3 MU	129	$self->conn();
2b5e63d1	130	eval { $self->{'svn'}->get_file($path,$rev,$fh); };
eaf718f3	131	};
8cd4177d	132	return () if $@ and $@ !~ /Attempted to get checksum/;
2b5e63d1	133	die $@ if $@;
eaf718f3 MU	134	close ($fh);
	135
	136	return ($name, $res);
	137	}
	138
	139
	140	package main;
	141
	142	my $svn = SVNconn->new($cvs_tree);
	143
	144
	145	sub pdate($) {
	146	my($d) = @_;
	147	$d =~ m#(\d\d\d\d)-(\d\d)-(\d\d)T(\d\d):(\d\d):(\d\d)#
	148	or die "Unparseable date: $d\n";
	149	my $y=$1; $y-=1900 if $y>1900;
	150	return timegm($6\|\|0,$5,$4,$3,$2-1,$y);
	151	}
	152
eaf718f3 MU	153	sub getwd() {
	154	my $pwd = `pwd`;
	155	chomp $pwd;
	156	return $pwd;
	157	}
	158
	159
	160	sub get_headref($$) {
	161	my $name = shift;
	162	my $git_dir = shift;
	163	my $sha;
	164
	165	if (open(C,"$git_dir/refs/heads/$name")) {
	166	chomp($sha = <C>);
	167	close(C);
	168	length($sha) == 40
	169	or die "Cannot get head id for $name ($sha): $!\n";
	170	}
	171	return $sha;
	172	}
	173
	174
	175	-d $git_tree
	176	or mkdir($git_tree,0777)
	177	or die "Could not create $git_tree: $!";
	178	chdir($git_tree);
	179
	180	my $orig_branch = "";
	181	my $forward_master = 0;
	182	my %branches;
	183
	184	my $git_dir = $ENV{"GIT_DIR"} \|\| ".git";
	185	$git_dir = getwd()."/".$git_dir unless $git_dir =~ m#^/#;
	186	$ENV{"GIT_DIR"} = $git_dir;
	187	my $orig_git_index;
	188	$orig_git_index = $ENV{GIT_INDEX_FILE} if exists $ENV{GIT_INDEX_FILE};
	189	my ($git_ih, $git_index) = tempfile('gitXXXXXX', SUFFIX => '.idx',
	190	DIR => File::Spec->tmpdir());
	191	close ($git_ih);
	192	$ENV{GIT_INDEX_FILE} = $git_index;
	193	my $maxnum = 0;
	194	my $last_rev = "";
	195	my $last_branch;
7ee74a99	196	my $current_rev = $opt_s ? ($opt_s-1) : 0;
eaf718f3 MU	197	unless(-d $git_dir) {
	198	system("git-init-db");
	199	die "Cannot init the GIT db at $git_tree: $?\n" if $?;
	200	system("git-read-tree");
	201	die "Cannot init an empty tree: $?\n" if $?;
	202
	203	$last_branch = $opt_o;
	204	$orig_branch = "";
	205	} else {
	206	-f "$git_dir/refs/heads/$opt_o"
	207	or die "Branch '$opt_o' does not exist.\n".
	208	"Either use the correct '-o branch' option,\n".
	209	"or import to a new repository.\n";
	210
	211	-f "$git_dir/svn2git"
	212	or die "'$git_dir/svn2git' does not exist.\n".
	213	"You need that file for incremental imports.\n";
	214	$last_branch = basename(readlink("$git_dir/HEAD"));
	215	unless($last_branch) {
	216	warn "Cannot read the last branch name: $! -- assuming 'master'\n";
	217	$last_branch = "master";
	218	}
	219	$orig_branch = $last_branch;
	220	$last_rev = get_headref($orig_branch, $git_dir);
	221	if (-f "$git_dir/SVN2GIT_HEAD") {
	222	die <<EOM;
	223	SVN2GIT_HEAD exists.
	224	Make sure your working directory corresponds to HEAD and remove SVN2GIT_HEAD.
	225	You may need to run
	226
	227	git-read-tree -m -u SVN2GIT_HEAD HEAD
	228	EOM
	229	}
	230	system('cp', "$git_dir/HEAD", "$git_dir/SVN2GIT_HEAD");
	231
	232	$forward_master =
	233	$opt_o ne 'master' && -f "$git_dir/refs/heads/master" &&
	234	system('cmp', '-s', "$git_dir/refs/heads/master",
	235	"$git_dir/refs/heads/$opt_o") == 0;
	236
	237	# populate index
	238	system('git-read-tree', $last_rev);
	239	die "read-tree failed: $?\n" if $?;
	240
	241	# Get the last import timestamps
	242	open my $B,"<", "$git_dir/svn2git";
	243	while(<$B>) {
	244	chomp;
	245	my($num,$branch,$ref) = split;
	246	$branches{$branch}{$num} = $ref;
	247	$branches{$branch}{"LAST"} = $ref;
e7e477df	248	$current_rev = $num if $current_rev < $num;
eaf718f3 MU	249	}
	250	close($B);
	251	}
	252	-d $git_dir
	253	or die "Could not create git subdir ($git_dir).\n";
	254
	255	open BRANCHES,">>", "$git_dir/svn2git";
	256
	257
	258	## cvsps output:
	259	#---------------------
	260	#PatchSet 314
	261	#Date: 1999/09/18 13:03:59
	262	#Author: wkoch
	263	#Branch: STABLE-BRANCH-1-0
	264	#Ancestor branch: HEAD
	265	#Tag: (none)
	266	#Log:
	267	# See ChangeLog: Sat Sep 18 13:03:28 CEST 1999 Werner Koch
	268	#Members:
	269	# README:1.57->1.57.2.1
	270	# VERSION:1.96->1.96.2.1
	271	#
	272	#---------------------
	273
	274	my $state = 0;
	275
	276	sub get_file($$$) {
	277	my($rev,$branch,$path) = @_;
	278
	279	# revert split_path(), below
	280	my $svnpath;
	281	$path = "" if $path eq "/"; # this should not happen, but ...
	282	if($branch eq "/") {
	283	$svnpath = "/$trunk_name/$path";
	284	} elsif($branch =~ m#^/#) {
	285	$svnpath = "/$tag_name$branch/$path";
	286	} else {
	287	$svnpath = "/$branch_name/$branch/$path";
	288	}
	289
	290	# now get it
2b5e63d1 MU	291	my ($name, $res) = eval { $svn->file($svnpath,$rev); };
2b5e63d1 MU	292	return () unless defined $name;
eaf718f3	293
6d281217	294	open my $F, '-\|', "git-hash-object", "-w", $name
eaf718f3 MU	295	or die "Cannot create object: $!\n";
	296	my $sha = <$F>;
	297	chomp $sha;
	298	close $F;
eaf718f3 MU	299	my $mode = "0644"; # SV does not seem to store any file modes
	300	return [$mode, $sha, $path];
	301	}
	302
	303	sub split_path($$) {
	304	my($rev,$path) = @_;
	305	my $branch;
	306
	307	if($path =~ s#^/\Q$tag_name\E/([^/]+)/?##) {
	308	$branch = "/$1";
	309	} elsif($path =~ s#^/\Q$trunk_name\E/?##) {
	310	$branch = "/";
	311	} elsif($path =~ s#^/\Q$branch_name\E/([^/]+)/?##) {
	312	$branch = $1;
	313	} else {
	314	print STDERR "$rev: Unrecognized path: $path\n";
	315	return ()
	316	}
	317	$path = "/" if $path eq "";
	318	return ($branch,$path);
	319	}
	320
	321	sub commit {
	322	my($branch, $changed_paths, $revision, $author, $date, $message) = @_;
	323	my($author_name,$author_email,$dest);
	324	my(@old,@new);
	325
2b5e63d1 MU	326	if (not defined $author) {
	327	$author_name = $author_email = "unknown";
	328	} elsif ($author =~ /^(.?)\s+<(.)>$/) {
eaf718f3 MU	329	($author_name, $author_email) = ($1, $2);
	330	} else {
	331	$author =~ s/^<(.*)>$/$1/;
	332	$author_name = $author_email = $author;
	333	}
	334	$date = pdate($date);
	335
	336	my $tag;
	337	my $parent;
	338	if($branch eq "/") { # trunk
	339	$parent = $opt_o;
	340	} elsif($branch =~ m#^/(.+)#) { # tag
	341	$tag = 1;
	342	$parent = $1;
	343	} else { # "normal" branch
	344	# nothing to do
	345	$parent = $branch;
	346	}
	347	$dest = $parent;
	348
	349	my $prev = $changed_paths->{"/"};
f0daa628	350	if($prev and $prev->[0] eq "A") {
eaf718f3	351	delete $changed_paths->{"/"};
f0daa628	352	my $oldpath = $prev->[1];
eaf718f3 MU	353	my $rev;
	354	if(defined $oldpath) {
	355	my $p;
	356	($parent,$p) = split_path($revision,$oldpath);
	357	if($parent eq "/") {
	358	$parent = $opt_o;
	359	} else {
	360	$parent =~ s#^/##; # if it's a tag
	361	}
	362	} else {
	363	$parent = undef;
	364	}
	365	}
	366
	367	my $rev;
7ee74a99	368	if($revision > $opt_s and defined $parent) {
eaf718f3 MU	369	open(H,"git-rev-parse --verify $parent \|");
	370	$rev = <H>;
	371	close(H) or do {
	372	print STDERR "$revision: cannot find commit '$parent'!\n";
	373	return;
	374	};
	375	chop $rev;
	376	if(length($rev) != 40) {
	377	print STDERR "$revision: cannot find commit '$parent'!\n";
	378	return;
	379	}
	380	$rev = $branches{($parent eq $opt_o) ? "/" : $parent}{"LAST"};
7ee74a99	381	if($revision != $opt_s and not $rev) {
eaf718f3 MU	382	print STDERR "$revision: do not know ancestor for '$parent'!\n";
	383	return;
	384	}
	385	} else {
	386	$rev = undef;
	387	}
	388
f0daa628	389	# if($prev and $prev->[0] eq "A") {
eaf718f3 MU	390	# if(not $tag) {
	391	# unless(open(H,"> $git_dir/refs/heads/$branch")) {
	392	# print STDERR "$revision: Could not create branch $branch: $!\n";
	393	# $state=11;
	394	# next;
	395	# }
	396	# print H "$rev\n"
	397	# or die "Could not write branch $branch: $!";
	398	# close(H)
	399	# or die "Could not write branch $branch: $!";
	400	# }
	401	# }
	402	if(not defined $rev) {
	403	unlink($git_index);
	404	} elsif ($rev ne $last_rev) {
	405	print "Switching from $last_rev to $rev ($branch)\n" if $opt_v;
	406	system("git-read-tree", $rev);
	407	die "read-tree failed for $rev: $?\n" if $?;
	408	$last_rev = $rev;
	409	}
	410
bf267d99 MU	411	my $cid;
	412	if($tag and not %$changed_paths) {
	413	$cid = $rev;
	414	} else {
	415	while(my($path,$action) = each %$changed_paths) {
	416	if ($action->[0] eq "A") {
	417	my $f = get_file($revision,$branch,$path);
	418	push(@new,$f) if $f;
	419	} elsif ($action->[0] eq "D") {
	420	push(@old,$path);
	421	} elsif ($action->[0] eq "M") {
	422	my $f = get_file($revision,$branch,$path);
	423	push(@new,$f) if $f;
	424	} elsif ($action->[0] eq "R") {
	425	# refer to a file/tree in an earlier commit
	426	push(@old,$path); # remove any old stuff
	427
	428	# ... and add any new stuff
	429	my($b,$p) = split_path($revision,$action->[1]);
	430	open my $F,"-\|","git-ls-tree","-r","-z", $branches{$b}{$action->[2]}, $p;
	431	local $/ = '\0';
	432	while(<$F>) {
	433	chomp;
	434	my($m,$p) = split(/\t/,$_,2);
	435	my($mode,$type,$sha1) = split(/ /,$m);
	436	next if $type ne "blob";
	437	push(@new,[$mode,$sha1,$p]);
	438	}
	439	} else {
	440	die "$revision: unknown action '".$action->[0]."' for $path\n";
	441	}
	442	}
	443
	444	if(@old) {
	445	open my $F, "-\|", "git-ls-files", "-z", @old or die $!;
	446	@old = ();
eaf718f3 MU	447	local $/ = '\0';
	448	while(<$F>) {
	449	chomp;
bf267d99 MU	450	push(@old,$_);
	451	}
	452	close($F);
	453
	454	while(@old) {
	455	my @o2;
	456	if(@old > 55) {
	457	@o2 = splice(@old,0,50);
	458	} else {
	459	@o2 = @old;
	460	@old = ();
	461	}
	462	system("git-update-index","--force-remove","--",@o2);
	463	die "Cannot remove files: $?\n" if $?;
eaf718f3	464	}
eaf718f3	465	}
bf267d99 MU	466	while(@new) {
	467	my @n2;
	468	if(@new > 12) {
	469	@n2 = splice(@new,0,10);
eaf718f3	470	} else {
bf267d99 MU	471	@n2 = @new;
bf267d99 MU	472	@new = ();
eaf718f3	473	}
bf267d99 MU	474	system("git-update-index","--add",
	475	(map { ('--cacheinfo', @$_) } @n2));
	476	die "Cannot add files: $?\n" if $?;
eaf718f3	477	}
eaf718f3	478
bf267d99 MU	479	my $pid = open(C,"-\|");
	480	die "Cannot fork: $!" unless defined $pid;
	481	unless($pid) {
	482	exec("git-write-tree");
	483	die "Cannot exec git-write-tree: $!\n";
eaf718f3	484	}
bf267d99 MU	485	chomp(my $tree = <C>);
	486	length($tree) == 40
	487	or die "Cannot get tree id ($tree): $!\n";
	488	close(C)
	489	or die "Error running git-write-tree: $?\n";
	490	print "Tree ID $tree\n" if $opt_v;
	491
	492	my $pr = IO::Pipe->new() or die "Cannot open pipe: $!\n";
	493	my $pw = IO::Pipe->new() or die "Cannot open pipe: $!\n";
	494	$pid = fork();
	495	die "Fork: $!\n" unless defined $pid;
	496	unless($pid) {
	497	$pr->writer();
	498	$pw->reader();
	499	open(OUT,">&STDOUT");
	500	dup2($pw->fileno(),0);
	501	dup2($pr->fileno(),1);
	502	$pr->close();
	503	$pw->close();
	504
	505	my @par = ();
	506	@par = ("-p",$rev) if defined $rev;
	507
	508	# loose detection of merges
	509	# based on the commit msg
	510	foreach my $rx (@mergerx) {
	511	if ($message =~ $rx) {
	512	my $mparent = $1;
	513	if ($mparent eq 'HEAD') { $mparent = $opt_o };
	514	if ( -e "$git_dir/refs/heads/$mparent") {
	515	$mparent = get_headref($mparent, $git_dir);
	516	push @par, '-p', $mparent;
	517	print OUT "Merge parent branch: $mparent\n" if $opt_v;
	518	}
	519	}
	520	}
eaf718f3	521
bf267d99 MU	522	exec("env",
	523	"GIT_AUTHOR_NAME=$author_name",
	524	"GIT_AUTHOR_EMAIL=$author_email",
	525	"GIT_AUTHOR_DATE=".strftime("+0000 %Y-%m-%d %H:%M:%S",gmtime($date)),
	526	"GIT_COMMITTER_NAME=$author_name",
	527	"GIT_COMMITTER_EMAIL=$author_email",
	528	"GIT_COMMITTER_DATE=".strftime("+0000 %Y-%m-%d %H:%M:%S",gmtime($date)),
	529	"git-commit-tree", $tree,@par);
	530	die "Cannot exec git-commit-tree: $!\n";
	531	}
	532	$pw->writer();
	533	$pr->reader();
eaf718f3	534
bf267d99	535	$message =~ s/[\s\n]+\z//;
eaf718f3	536
bf267d99 MU	537	print $pw "$message\n"
	538	or die "Error writing to git-commit-tree: $!\n";
	539	$pw->close();
eaf718f3	540
bf267d99 MU	541	print "Committed change $revision:$branch ".strftime("%Y-%m-%d %H:%M:%S",gmtime($date)).")\n" if $opt_v;
	542	chomp($cid = <$pr>);
	543	length($cid) == 40
	544	or die "Cannot get commit id ($cid): $!\n";
	545	print "Commit ID $cid\n" if $opt_v;
	546	$pr->close();
eaf718f3	547
bf267d99 MU	548	waitpid($pid,0);
	549	die "Error running git-commit-tree: $?\n" if $?;
	550	}
eaf718f3	551
f02b3eba MU	552	if(not defined $dest) {
	553	print "... no known parent\n" if $opt_v;
	554	} elsif(not $tag) {
eaf718f3 MU	555	print "Writing to refs/heads/$dest\n" if $opt_v;
	556	open(C,">$git_dir/refs/heads/$dest") and
	557	print C ("$cid\n") and
	558	close(C)
	559	or die "Cannot write branch $dest for update: $!\n";
eaf718f3	560	}
eaf718f3 MU	561
	562	if($tag) {
	563	my($in, $out) = ('','');
	564	$last_rev = "-" if %$changed_paths;
	565	# the tag was 'complex', i.e. did not refer to a "real" revision
	566
f02b3eba	567	$dest =~ tr/_/\./ if $opt_u;
eaf718f3 MU	568
	569	my $pid = open2($in, $out, 'git-mktag');
	570	print $out ("object $cid\n".
	571	"type commit\n".
f02b3eba	572	"tag $dest\n".
eaf718f3 MU	573	"tagger $author_name <$author_email>\n") and
eaf718f3 MU	574	close($out)
f02b3eba	575	or die "Cannot create tag object $dest: $!\n";
eaf718f3 MU	576
	577	my $tagobj = <$in>;
	578	chomp $tagobj;
	579
	580	if ( !close($in) or waitpid($pid, 0) != $pid or
	581	$? != 0 or $tagobj !~ /^[0123456789abcdef]{40}$/ ) {
f02b3eba	582	die "Cannot create tag object $dest: $!\n";
eaf718f3	583	}
eaf718f3	584
f02b3eba MU	585	open(C,">$git_dir/refs/tags/$dest") and
f02b3eba MU	586	print C ("$tagobj\n") and
eaf718f3	587	close(C)
f02b3eba	588	or die "Cannot create tag $branch: $!\n";
eaf718f3	589
f02b3eba	590	print "Created tag '$dest' on '$branch'\n" if $opt_v;
eaf718f3	591	}
e7e477df MU	592	$branches{$branch}{"LAST"} = $cid;
	593	$branches{$branch}{$revision} = $cid;
	594	$last_rev = $cid;
	595	print BRANCHES "$revision $branch $cid\n";
	596	print "DONE: $revision $dest $cid\n" if $opt_v;
eaf718f3 MU	597	}
	598
	599	my ($changed_paths, $revision, $author, $date, $message, $pool) = @_;
	600	sub _commit_all {
	601	($changed_paths, $revision, $author, $date, $message, $pool) = @_;
f0daa628 MU	602	my %p;
	603	while(my($path,$action) = each %$changed_paths) {
	604	$p{$path} = [ $action->action,$action->copyfrom_path, $action->copyfrom_rev ];
	605	}
	606	$changed_paths = \%p;
eaf718f3	607	}
f0daa628	608
eaf718f3 MU	609	sub commit_all {
	610	my %done;
	611	my @col;
	612	my $pref;
	613	my $branch;
	614
	615	while(my($path,$action) = each %$changed_paths) {
	616	($branch,$path) = split_path($revision,$path);
	617	next if not defined $branch;
	618	$done{$branch}{$path} = $action;
	619	}
	620	while(($branch,$changed_paths) = each %done) {
	621	commit($branch, $changed_paths, $revision, $author, $date, $message);
	622	}
	623	}
	624
	625	while(++$current_rev < $svn->{'maxrev'}) {
	626	$svn->{'svn'}->get_log("/",$current_rev,$current_rev,$current_rev,1,1,\&_commit_all,"");
	627	commit_all();
3ef378a6 MU	628	if($opt_l and not --$opt_l) {
	629	print STDERR "Exiting due to a memory leak. Repeat, please.\n";
	630	last;
	631	}
eaf718f3 MU	632	}
	633
	634
	635	unlink($git_index);
	636
	637	if (defined $orig_git_index) {
	638	$ENV{GIT_INDEX_FILE} = $orig_git_index;
	639	} else {
	640	delete $ENV{GIT_INDEX_FILE};
	641	}
	642
	643	# Now switch back to the branch we were in before all of this happened
	644	if($orig_branch) {
3ef378a6	645	print "DONE\n" if $opt_v and (not defined $opt_l or $opt_l > 0);
eaf718f3 MU	646	system("cp","$git_dir/refs/heads/$opt_o","$git_dir/refs/heads/master")
	647	if $forward_master;
	648	unless ($opt_i) {
	649	system('git-read-tree', '-m', '-u', 'SVN2GIT_HEAD', 'HEAD');
	650	die "read-tree failed: $?\n" if $?;
	651	}
	652	} else {
	653	$orig_branch = "master";
3ef378a6	654	print "DONE; creating $orig_branch branch\n" if $opt_v and (not defined $opt_l or $opt_l > 0);
eaf718f3 MU	655	system("cp","$git_dir/refs/heads/$opt_o","$git_dir/refs/heads/master")
	656	unless -f "$git_dir/refs/heads/master";
	657	unlink("$git_dir/HEAD");
	658	symlink("refs/heads/$orig_branch","$git_dir/HEAD");
	659	unless ($opt_i) {
	660	system('git checkout');
	661	die "checkout failed: $?\n" if $?;
	662	}
	663	}
	664	unlink("$git_dir/SVN2GIT_HEAD");
	665	close(BRANCHES);